Our client, a car insurance company, is now looking for an Analytics Engineer to take ownership of analytics, reporting, and BI dashboards for a growing AI-driven team within a leading technology organization.

The team produces high volumes of production data, including model predictions, chatbot interactions, and operational automation. Currently, there is a need for a dedicated analyst to monitor, analyze, and visualize this data to provide actionable insights across the organization.

Responsibilities:

Dashboard ownership (primary focus):

• Design, build, and maintain Looker dashboards that track ML model performance across approximately 12 production models

• Own the "what to show" decision for each dashboard — choose appropriate metrics, time windows, and granularity independently

• Translate model performance data into business terms for stakeholders — enabling decisions like "is it time to retrain?" or "is there a feature data problem?" rather than reporting raw statistical outputs

Data modeling and transformation:

• Extend and maintain existing DBT incremental models that join predictions to actuals with time-lag offsets (e.g., predictions from day T joined to actuals from day T+15)

• Apply DBT materialization strategies (incremental, table, view) and manage full refreshes after schema or logic changes

Monitoring and collaboration:

• Design alerting logic to flag model degradation or abnormal prediction patterns

• Collaborate with ML engineers to understand each model's prediction logic and define what healthy performance looks like

• Respond to ad-hoc analytical questions from ML engineers, product, and leadership

Requirements:

Core Requirements:

Preferred Qualifications:

• Experience with model monitoring concepts (data drift, feature drift, concept drift)

• Insurance or fintech domain experience (conversion funnels, policy lifecycle, claims)

• Experience working embedded within AI/ML engineering teams

• Python for ad-hoc analysis and automation